Discussing zebrafish genetic methods

This blog was originally set up to help plan a workshop held at the 2007 SCZI. During this workshop, it was clear that while there are some powerful new advances in genetic methods for the zebrafish, there are important questions about some existing methods, and several methods (notably homologous recombination and RNAi) which would be great for the field, but do not yet exist.

We've left this blog up as a forum in which people can discuss some of these methods. Below are several blog entries describing particular methods. Please add questions and comments under the appropriate heading.

--Chi-Bin Chien, Koichi Kawakami, Todd Evans, Hazel Sive

Monday, January 22, 2007

Topics suggested for the workshop

This was a place to suggest topics for the workshop. Since this thread is now obsolete, please don't comment here.

Community resources

Here are some resources that might be useful to develop as a community:

• ORFeome libraries (coding sequences in Gateway entry vectors)
• enhancer libraries (enhancers or enhancer/promoters in Gateway entry vectors)
• Gal4 enhancer trap lines
• tissue-specific Cre lines

We note that distribution of fish lines will be laborious. ZIRC is working on a model where they visit the screening lab and cryopreserve sperm on-site.

It would be very useful to have the community's comments on the utility of ORFeome, enhancerome, or promoterome libraries.

RNAi methods

RNAi is used routinely in nematodes and flies, while siRNA methods have become prevalent in mammalian cell culture as well as in mice. There is no obvious reason why it should not be possible to use siRNA methods in zebrafish. This would be especially useful to achieve tissue-specific knockdown.

Anecdotally, many labs have tried various approaches to RNAi, though apparently without any solid success to date. Hazel Sive's lab is trying an siRNA-based method, and Hazel has offered to lead an online discussion about this. She is planning to set up a blog for this soon, and we will post the URL here.

Homologous recombination

Homologous recombination is clearly the one genetic method where zebrafish are furthest behind mice as a model system. Taking mice as a model, it would be very useful to have a pluripotent cell line that would allow transfection, selection for rare homologous recombination events, and then generation of a complete fish.

The Collodi lab has been working for a while on developing embryonic stem cells; other ideas have been to generate embryonic germ cell cultures or to use somatic cell nuclear transplantation (cloning) to allow use of fibroblasts or other existing cultures.

Another potential method could be to generate targeted double-strand breaks in an embryo (e.g. using genetically engineered zinc-finger nucleases). Jim Amatruda's, Masahako Hibi's, and Joseph Yost's lab are all trying this approach.

Conditional or tissue-specific expression

In flies, the Gal4/UAS system is invaluable for driving tissue-specific expression of transgenes. In mice, the Cre/Lox system and more recently, Cre/Lox combined with Tet on/off or regulatable Cre (e.g. CreER) has played a similar role.

Several labs are conducting Gal4-enhancer trap screens, including Koichi Kawakami's, Herwig Baier's, Marnie Halpern's, and Steve Johnson's. Papers describing pilot screens are in preparation, and most labs seem to be planning to post images of their insertions and make the lines available to the community.

There has recently been published work from several labs, including Tom Look's, on Cre/Lox. Laura Bally-Cuif's lab has had some success with CreERT.

It is not clear how well Cre works--for instance, does recombination recur in every single cell that expresses Cre?

Transient expression methods

There have been published results showing in vivo electroporation, lipofection, and viral infection in zebrafish embryos; the electroporation methods seem likely to be the most generally applicable. Jon Clarke and Steve Wilson's labs have shown very nice results with single-cell in vivo electroporation.

Global heat-shock of embryos and larvae is becoming very widely used as a method for expressing gene products at particular times (for instance to do late rescue of a mutant).

The Chien lab has also developed a very low-tech, reliable method for local heat-shock induction (using local heating with a modified soldering iron rather than sublethal laser illumination). A manuscript is in preparation; if you email Chi-Bin after about 1 March, we should be able to send you a draft.

Gateway-based systems for building expression vectors

Building complex expression vectors with components such as large regulatory elements, fluorescent protein fusions, transgenesis markers, or bicistronic (IRES) markers can be extremely time-consuming. The labs of Nathan Lawson and Chi-Bin Chien have both built systems using multisite Gateway technology, which allows quick construction of expression vectors in a Tol2 backbone. Given Gateway-based "ORFeome" libraries, it should also become possible to do large-scale misexpression screens.

We have already built many entry vectors that allow introduction of ubiquitous promoters (beta-actin, hsp70, CMV/SP6), fluorescent protein markers (EGFP, mCherry, and nuclear and membrane-tagged versions of both), as well as N- and C-terminal fusions, and several IRES tags. Nathan and Chi-Bin made a short presentation on these constructs, and distributed clones at the meeting.

For documentation and more information, see:

http://chien.neuro.utah.edu/tol2kitwiki

or
http://lawsonlab.umassmed.edu

We also know that Jochen Wittbrodt's lab has built a similar system using I-SceI vectors, while Shannon Fisher and Michael Nonet have built systems using "classic" [not multisite] Gateway.